Evaluation of the reliability of pulse oximetry, at different attachment sites, to detect hypoxaemia in immobilized impala (Aepyceros melampus) ...

Page created by Erin Gibson
 
CONTINUE READING
Evaluation of the reliability of pulse oximetry, at different attachment sites, to detect hypoxaemia in immobilized impala (Aepyceros melampus) ...
Veterinary Anaesthesia and Analgesia 2020, 47, 323e333                                                    https://doi.org/10.1016/j.vaa.2019.08.051

RESEARCH PAPER

Evaluation of the reliability of pulse oximetry, at
different attachment sites, to detect hypoxaemia in
immobilized impala (Aepyceros melampus)

Thembeka K Mtetwaa,b, Gareth E Zeilerc, Liesel Laubscherd,e, Silke Pfitzerf,g & Leith CR Meyera,b
a
    Department of Paraclinical Sciences, Faculty of Veterinary Science, University of Pretoria, Onderstepoort, South Africa
b
    Centre for Veterinary Wildlife Studies, Faculty of Veterinary Science, University of Pretoria, Onderstepoort, South
Africa
c
 Department of Companion Animal Clinical Studies, Faculty of Veterinary Science, University of Pretoria,
Onderstepoort, South Africa
d
    Wildlife Pharmaceuticals, Rocky Drift, White River, South Africa
e
    Department of Animal Science, Faculty of Agriculture, University of Stellenbosch, Matieland, South Africa
f
Murdoch University, School of Veterinary and Life Sciences, Murdoch University, Perth, Western Australia, Australia
g
    Wildlife Vets, Ngongoni Farm, Tipperary, South Africa

Correspondence: Thembeka K Mtetwa, Department of Paraclinical Sciences, Faculty of Veterinary Science, University of Pretoria, P. Bag X04, Onderstepoort,
Gauteng, 0110, South Africa. E-mail: thembeka.mtetwa@up.ac.za

Abstract                                                                       precision ¼ 3, Arms ¼ 4). The ear, cheek and rectal probes
                                                                               failed to give accurate or precise readings (ear: bias ¼ 4,
Objective Evaluation of the reliability of pulse oximetry at
                                                                               precision ¼ 14, Arms ¼ 15; cheek: bias ¼ 12, precision ¼
four different attachment sites compared to haemoglobin
                                                                               11, Arms ¼ 16; and rectum: bias ¼ 5, precision ¼ 12,
oxygen saturation measured by a co-oximeter and calcu-
                                                                               Arms ¼ 13).
lated by a blood gas analyser in immobilized impala.
                                                                               Conclusions and clinical relevance In order to obtain ac-
Study design Randomized crossover study.
                                                                               curate and precise pulse oximetry readings in immobilized
Animals A total of 16 female impala.                                           impala, probes must be placed under the tail and SaO2 must
                                                                               be above 90%. Since SaO2 values are usually low in
Methods Impala were immobilized with etorphine or thia-
                                                                               immobilized impala, pulse oximeter readings should be
fentanil alone, or etorphine in combination with a novel
                                                                               interpreted with caution.
drug. Once immobilized, arterial blood samples were
collected at 5 minute intervals for 30 minutes. Then oxygen
was insufflated (5 L minute1) intranasally at 40 minutes                       Keywords immobilization, impala, oxygen saturation,
and additional samples were collected. A blood gas analyser                    pulse oximetry.
was used to measure the arterial partial pressure of oxygen
and calculate the oxygen haemoglobin saturation (cSaO2); a
                                                                               Introduction
co-oximeter was used to measure the oxygen haemoglobin
saturation (SaO2) in arterial blood. Pulse oximeter probes                     Chemical immobilization is an essential tool for wildlife con-
were attached: under the tail, to the pinna (ear) and buccal                   servation and management worldwide. It is commonly used in
mucosa (cheek) and inside the rectum. Pulse oximeter                           impala, mainly, but not exclusively, for translocation and dis-
readings [peripheral oxygen haemoglobin saturation (SpO2)                      ease surveillance purposes. To immobilize impala, potent opi-
and pulse quality] were recorded at each site and compared                     oids (such as etorphine or thiafentanil) are primarily used.
with SaO2 and cSaO2 using Bland-Altman and accuracy of                         Impala are particularly sensitive to the adverse effects of these
the area root mean squares (Arms) methods to determine the                     drugs and deaths often occur when they are captured (Meyer
efficacy. P value < 0.05 was considered significant.                             et al. 2008; Zeiler & Meyer 2017a). These opioids impair
                                                                               ventilatory function by causing ventilatory depression, intra-
Results Pulse quality was ‘good’ at each attachment site.
                                                                               pulmonary shunting, ventilation-perfusion mismatch and
SpO2 measured under the tail was accurate and precise but
                                                                               probable impedance to oxygen diffusion, ultimately leading to
only when SaO2 values were above 90% (bias ¼ 3,

                                                                                                                                                    323
Reliability of pulse oximetry in impala TK Mtetwa et al.

severe hypoxaemia (Zeiler & Meyer 2017a,b). Untreated                          Materials and methods
hypoxaemia may cause tissue hypoxia, organ failure and death
(Fahlman 2014). Therefore, it is important that blood                          Animals
oxygenation is continuously and accurately monitored during
                                                                               The study was approved by the Animal Ethics Committee of the
immobilization to ensure that hypoxaemia is detected and
                                                                               University of Pretoria (V035-17) and was conducted at
treated early to guarantee the welfare and survival of an
                                                                               Ngongoni Farm (25 31’25.2”S, 31 06’50.8”E). A total of 16
animal.
                                                                               free-ranging adult female impala (Aepyceros melampus),
   Currently, co-oximeters and blood gas machines are
                                                                               temporarily housed together in a 6  8 m outdoor enclosure
frequently used for the detection of low arterial oxygen hae-
                                                                               (boma), were used. The impala [body weight 34.1 ± 5.2 kg d
moglobin saturation (SaO2) or oxygen partial pressures or both
                                                                               mean ± standard deviation (SD)] were habituated to the
(Wong et al. 2011; Bardell et al. 2017). However, these devices
                                                                               housing conditions 14 days prior to the study. Lucerne (Med-
do not give continuous readings, they can be costly and the
                                                                               icago sativa) and water were supplied ad libitum with game
arterial samples they require can be difficult to obtain in some
                                                                               pellets supplemented as needed. For exclusion, animals were
species, including immobilized impala (Proulx 1999).
                                                                               checked for overt signs of cardio-respiratory and other diseases
Furthermore, these devices are not always practical to use in
                                                                               at the first immobilization.
the field (Proulx 1999). Pulse oximetry (SpO2), which gives an
indirect peripheral measure of SaO2 (Pedersen et al. 2009), has
                                                                               Immobilization
become a popular method used for the early detection of
hypoxaemia in human and veterinary medicine (van de Louw                       Data for this study were collected during an unrelated study
et al. 2001; Young et al. 2002; Giguere et al. 2014; Haymerle                  (see Pfitzer et al. 2019) which aimed to assess the dose-
et al. 2016). It is widely used in the hospital and field setting               response effect of the serotonin R-enantiomer of 8-hydroxy-
because pulse oximetry devices are cheap, portable, easy to                    2-(di-n-propylamino) tetralin drug on opioid-induced respira-
apply and use, and give continuous noninvasive measures.                       tory depression. This ‘novel drug study’ used a randomized
Their use and effectiveness have been determined and well                      crossover study design, in which the impala were immobilized
documented in humans (Lee et al. 2000; van de Louw et al.                      (one at a time) on six different occasions. In each trial the
2001) and domestic animals (Chaffin et al. 1996; Mathews                        impalas were darted with either etorphine alone [0.09 mg
et al. 2003; Giguere et al. 2014; Grubb & Anderson 2017;                       kg1, Captivon, Wildlife Pharmaceuticals (Pty) Ltd., South
Reiners et al. 2018). However, although they are commonly                      Africa], thiafentanil alone [0.09 mg kg1, Thianil, Wildlife
used during the immobilization of wildlife (Martin-Jurado et al.               Pharmaceuticals (Pty) Ltd.] or etorphine (0.09 mg kg1) in
2011; Haymerle et al. 2016), their accuracy and utility have                   combination with different doses of the novel drug, using a 1.5
not been adequately determined. Therefore, the validity of their               mL dart (P-type Pneudarts with 19 mm long needle; DAN-
use for this purpose is still questionable.                                    INJECT, International S.A, South Africa) projected into a
   Studies in human and domesticated animals show that the                     muscle mass of the pelvic girdle by a carbon dioxide (CO2) gas
accuracy of pulse oximetry is dependent upon probe placement                   powered rifle [230 kPa: X-Caliber dart gun; Wildlife Pharma-
(Chaffin et al. 1996; Lee et al. 2000; van de Louw et al. 2001,                 ceuticals (Pty) Ltd.]. The dart rifle operator stood on an
Mathews et al. 2003; Giguere et al. 2014). In order to obtain a                elevated walkway above the boma wall and darted the impala
pulse oximeter reading, veterinarians working with immobi-                     over a distance ranging from 5 to 12 m.
lized wildlife place probes at different sites including the ear, lip,
rectum, tongue, vulva and the tip of the penis (Proulx 1999).                  Experimental procedures
However, whether the readings from these sites are accurate
                                                                               Following immobilization, the impala were blindfolded and
and precise still needs to be elucidated. Although several
                                                                               carried to an under roof workstation and placed in sternal re-
studies have previously compared SpO2 readings at different
                                                                               cumbency on a table. A peripheral artery (either arteriae
attachment sites in domestic animals (Chaffin et al. 1996;
                                                                               auricularis posterior or arteriae digitales palmares communes) was
Mathews et al. 2003; Giguere et al. 2014), no such studies
                                                                               aseptically cannulated (22 gauge; Jelco Smiths Medical, UK) for
have been performed in immobilized wildlife.
                                                                               arterial blood sampling. A total of four pulse oximeters (Nonin
   Therefore, this study aimed to establish, in immobilized
                                                                               PalmSat 2500 A; Kyron Laboratories (Pty) Ltd., South Africa)
impala, whether pulse oximetry is a reliable method for the
                                                                               were connected, using four probes [Kyron Laboratories (Pty)
determination of SaO2. The study also aimed to determine
                                                                               Ltd.] at different locations on the impala, as follows (Fig. 1):
which attachment site of the pulse oximeter probe provided the
most accurate SpO2 measurement. We hypothesized that pulse                     1) Taildreflectance probe (2000T Transflectance probe; Kyron
oximetry can be used to accurately measure SaO2 in immo-                          Laboratories (Pty) Ltd.) placed and secured gently, using adhe-
bilized impala under field conditions.                                             sive tape (25 mm ElastoPlast Tape; BSN Medical (Pty) Ltd., South

324       © 2020 Association of Veterinary Anaesthetists and American College of Veterinary Anesthesia and Analgesia. Published by Elsevier Ltd. All rights
                                                                                                                                reserved., 47, 323e333
Reliability of pulse oximetry in impala TK Mtetwa et al.

   Africa), on unpigmented skin on the ventral aspect of the tail                sensor and 8000R-reflectance sensor), which are compatible
   base within 20e30 mm from the root of the tail (Fig. 1a & a1).                with each other (https://www.nonin.com/wp-content/
2) Rectumdreflectance probe (2000T Transflectance probe)                           uploads/2018/09/Sensor-Compatibility-Guide.pdf). The de-
   inserted approximately 10 mm into the rectum and secured in
                                                                                 vice’s set averaging time for SpO2 readings is 10 seconds
   position by taping the connecting wire of the probe to the tail
   base using adhesive tape (Fig. 1b & b1).                                      (https://www.nonin.com/wp-content/uploads/Operators-
3) Eardtransmission probe (2000SL Lingual probe; Kyron Labo-                     Manual-2500A-Vet.pdf).
   ratories (Pty) Ltd.) clipped onto a shaved unpigmented area on                   In all the trials, 1 mL of arterial blood was collected in
   the caudal margin of the pinna (Fig. 1c & c1).                                heparinized plastic syringes (B Braun, Germany) and stored on
4) Cheekdreflectance probe (2000T Transflectance probe) secured                    ice at 5, 10, 15, 20 and 30 minutes from the time the impala
   to a holding clip and positioned inside the mouth and placed                  became recumbent. At 40 minutes of immobilization, seven
   against the buccal mucosa approximately 30 mm from the
                                                                                 impala were given etorphine alone or thiafentanil alone and
   commissure of the lip (Fig. 1d & d1).
                                                                                 were supplemented with oxygen for 10 minutes using a high-
   These sites were chosen based on practical application of the                 pressure oxygen cylinder [Ecomed Medical (Pty) Ltd, South
probes. The Nonin PalmSAT 2500 veterinary probes were used                       Africa]. The flowmeter was set at 5 L minute1, administered
in the study (2000T and 2000SL) because they are similar to                      intranasally via a nasal tube to increase the animal’s oxygen
the Nonin PalmSAT 2500 human probes (8000Q2-ear clip                             haemoglobin saturation. Thereafter, arterial samples were also

Figure 1 Placement of pulse oximeter probe at four different attachment sites. The tail (a) probe was a transflectance probe (a1). It was placed on
unpigmented skin at the base of the tail and secured with tape. The probe inserted into the rectum (b) was a transflectance probe (b1). It was placed
10 mm inside the rectum facing the mucosa next to the temperature probe and indicated by the black arrow. The lingual transmission probe (c1) was
clipped to the skin of the shaved margin of the ear (c) of the impala. The probe attached to the cheek (d) was a transflectance probe. It was attached to
a plastic clip with rubber bands (d1) and the plastic clip secured the probe adjacent to cheek the 30 mm from the commissure of the lips. White dotted
arrows represent internal probes placed on the mucosa and the white solid arrows are probes that are placed externally on the skin.
© 2020 Association of Veterinary Anaesthetists and American College of Veterinary Anesthesia and Analgesia. Published by Elsevier Ltd. All rights   325
reserved., 47, 323e333
Reliability of pulse oximetry in impala TK Mtetwa et al.

collected at random intervals during oxygen supplementation.                   random errordby calculating the SD of the differences), and
Peripheral oxygen haemoglobin saturation (SpO2), pulse rate                    limits of agreement (LOA) (calculated as the bias ± 1.96 SD) at
(beats minute1) and pulse quality data detected by the pulse                  95% confidence interval for the mean difference and LOA. A
oximeter were recorded in triplicate, within 30 seconds of                     frequency distribution test and column statistics were per-
collecting each arterial blood sample. Pulse quality (pulse                    formed to determine whether the mean differences between
signal strength) was indicated by a flashing coloured light                     the two variables (methods) were normally distributed.
emitting diode built into the pulse oximeter; green indicated                     The combined accuracy and precision of the compared
good, amber intermediate and red a poor quality. The pulse                     variables were determined by using the accuracy of the area
oximetry readings were recorded in the following order: 1) tail;               root mean square (Arms) (Batchelder & Raley 2007; Equation
2) rectum; 3) cheek; and 4) ear.                                               (1) in Appendix A). According to ISO 80601-2-61:2011
                                                                               guidelines, pulse oximetry is regarded as accurate and precise
Arterial blood analysis                                                        when the bias, precision and Arms are  ±3%,  3% and 
                                                                               ±4%, respectively (DIN EN, 2011).
After withdrawal, arterial blood samples were immediately
stored on ice and analyzed within 10 minutes after collection
                                                                               Data handling
using: 1) an EPOC blood gas analysis system and self-
calibrating BGEM3 test cards (Epocal Inc., ON, Canada); and                    The SpO2 data set was paired with the SaO2 and cSaO2 data
2) a daily calibrated Avoximeter 4000 co-oximeter with cu-                     sets to determine the total number of pairs available for anal-
vettes [Surgical Innovations (Pty) Ltd., South Africa)]. The                   ysis. Initially, all of the SpO2 measurements (all-data) were
blood gas analyser was used to measure the arterial partial                    used to calculate the mean of the triplicates for each data point
pressure of oxygen (PaO2) to calculate the arterial oxygen                     and these values were used for analyses using the methods
haemoglobin saturation (cSaO2) using a default equation                        described. Then, exclusion criteria were applied to the SpO2
programmed into the blood gas analyser by the manufacturer.                    data set to filter out poor pulse quality data points, thereby to
The co-oximeter was used to measure the fractional arterial                    obtain a good quality data set (pass-data). It was done by
oxygen haemoglobin saturation of the arterial blood (SaO2).                    excluding any triplicate data sets: 1) with red oximeter pulse
                                                                               quality light reading; and 2) where the pulse quality light
Statistical analysis                                                           indicated green or amber, but the SpO2 triplicate data had an
                                                                               SD of more than 3%.
For Bland-Altman method analysis to be performed at an a ¼
                                                                                  To determine the performance of the pulse oximeters at
0.05 and b ¼ 0.90, this study needed a minimum of 17 paired
                                                                               different SaO2 ranges, the SpO2 data were compared at the
data sets (i.e. SpO2 - SaO2, cSaO2 - SaO2 and SpO2 - cSaO2) to
                                                                               following different SaO2 or cSaO2 ranges: 0 e 100%, 70 e
have power when the expected mean of differences is 3%, ex-
                                                                               100% (range claimed by the manufacturer to be most accurate
pected SD of differences is 3% and the maximum allowed dif-
                                                                               for the pulse oximeter), < 70% (range claimed by the manu-
ference between methods is ±10%.
                                                                               facturer to be inaccurate for pulse oximetry), 70 e 79%, 80 e
   The SaO2 measured by the co-oximeter is accepted by the
                                                                               89% and 90 e 100% for ‘all-data’ and the ‘pass-data’ sets.
International Organisation for Standardisation (ISO) 80601-2-
                                                                                  Analyses were performed using commercially available
61:2011 guidelines as the gold-standard measurement of SaO2
                                                                               software MedCalc Version 19, MedCalc Software Ltd.,
to validate the use of pulse oximetry (DIN EN, 2011). The cSaO2
                                                                               (Belgium) and GraphPad Prism, Version 7, (GraphPad Soft-
calculated by the blood gas analyser is used as a ‘clinical stan-
                                                                               ware, CA, USA).
dard’ (Whitehair et al. 1990; Reiners et al. 2018). Both mea-
sures (SaO2 and cSaO2) were used to assess the accuracy and                    Results
utility of the pulse oximeters in this study. The Bland-Altman
method for multiple observations (Bland & Altman 2007) was                     Quality of data collected
used to determine the accuracy and precision of the measure-
ments recorded using the three monitors. More specifically: 1)                  All immobilization procedures were successful (Pfitzer et al.
how SpO2 measured by the pulse oximeters at each site                          2019). Overall, there were 167 paired SaO2 readings
compared with the SaO2; 2) how cSaO2 calculated by the blood                   measured by the co-oximeter and the SaO2 data ranged from
gas analyser compared with the measured SaO2; and 3) how                       16.3% to 99.3%. There were 194 paired cSaO2 readings
SpO2 measurements at each site compared with the cSaO2.                        calculated by the blood gas analyser and the data ranged from
   The Bland-Altman method calculates the bias [the measure                    15.5% to 99.9%. The SaO2 readings had a good agreement
of accuracydby calculating the mean difference between the                     (bias 1, LOA 5 to 7) with cSaO2 readings (Fig. 2).
two variables (methods) that are compared, i.e. SpO2 - SaO2,                      All probe sites gave good pulse quality readings, i.e. per-
cSaO2 - SaO2 and SpO2 - cSaO2], the precision (a measure of                    centage readings that had a good pulse qualitydgreen light,
326       © 2020 Association of Veterinary Anaesthetists and American College of Veterinary Anesthesia and Analgesia. Published by Elsevier Ltd. All rights
                                                                                                                                reserved., 47, 323e333
Reliability of pulse oximetry in impala TK Mtetwa et al.

(Tables 1 and 2), with the highest values obtained from the tail                                       readings, however, the Arms indicated that pulse oximetry was
and the lowest from the ear. The ear, the cheek and the rectum                                         accurate for ‘all-data’ and ‘pass-data’. Below these ranges,
had SpO2 readings indicating poor pulse quality (poor signal                                           pulse oximetry was inaccurate and imprecise. In the manu-
strengthdred light).                                                                                   facturer’s claimed performance range (70e100%) the read-
   Although the tail gave the best quality SpO2 readings, there                                        ings were accurate but imprecise (Table 4).
was a larger variability in the readings obtained from the tail, and
likewise from the ear (triplicate readings with SD>3%, Tables 1                                        Ear
and 2) compared with the other two sites (rectum and cheek).
                                                                                                       When compared with SaO2 (Tables 3 and 4), ‘all-data’
                                                                                                       readings were accurate but imprecise in all ranges except
Pulse oximeter performance at each site
                                                                                                       within the 70e79% and 90e100% ranges, where the
The mean difference between the two variables (methods) that                                           readings were both inaccurate and imprecise. These findings
were compared, i.e. SpO2eSaO2, cSaO2eSaO2 and SpO2ec-                                                  were similar to the ‘pass-data’. At the range of 70e100%, the
SaO2, were normally distributed.                                                                       pulse oximeter gave accurate but imprecise readings for ‘all-
                                                                                                       data’ and ‘pass-data’ (Fig. 3). When compared with cSaO2, the
Tail                                                                                                   ear probe was inaccurate and imprecise at all ranges.

Compared with SaO2 (Tables 3), the pulse oximeter had an
                                                                                                       Rectum
acceptable performance (i.e. it was accurate and precise) in the
SaO2 range of 90e100% for ‘all-data’. In the SaO2 range of                                             Within the SaO2 ranges of 80e89% and 90e100%, the pulse
80e89%, below this range (70e79% and
Reliability of pulse oximetry in impala TK Mtetwa et al.

                                                          a                                     b
                                                         40
                 Mean difference [SpO2-SaO2 (% )]

                                                         20

                                                          0

                                                        –20

                                                        –40

                                                          c                                     d
                                                         40
                    Mean difference [SpO2-cSaO2 (% )]

                                                         20

                                                          0

                                                        –20

                                                        –40
                                                                                                       Tail

                                                                                                                            Rectum
                                                                                                                  Ear

                                                                                                                                       Cheek
                                                              Tail

                                                                           Rectum
                                                                     Ear

                                                                                    Cheek

Figure 3 The performance of the peripheral oxygen haemoglobin saturation (SpO2; measured by pulse oximetry) compared with arterial oxygen
haemoglobin saturation (SaO2; measured by the co-oximeter) (graphs a and b) and arterial oxygen haemoglobin saturation (cSaO2; calculated by
blood gas analyser) (graphs c and d) in the range of 70 e 100% (manufacturer’s claimed performance range of the pulse oximeters) shown on the
y axis, at four probe attachment sites in 16 impala, shown on the x axis. a and c indicate ‘all-data’ sets (i.e. when exclusion criteria were not
applied) and b and d indicate “pass-data” sets (i.e. when exclusion criteria were applied). The data were analysed using the Bland-Altman method
for multiple comparisons per individual. The mean differences (biaseblack squares) (SpO2-SaO2; SpO2-cSaO2) and the limits of agreement [mean ±
1.96 standard deviation (SD) indicated by the error bars] are plotted for the four probe attachment sites. The black line at 0 indicates a bias of 0;
the closer the mean differences are to this line, the greater the accuracy of the pulse oximeter measurements and the smaller the limits of
agreement, the greater the precision of these measurements.

Cheek                                                                                       provided good pulse quality readings; however, the variability
                                                                                            of these readings was also high. The accuracy of pulse oximetry
When compared with SaO2 (Table 3), pulse oximetry was
                                                                                            was poor when the probes were placed on the ear, inside the
inaccurate and imprecise at all ranges except within the range
                                                                                            rectum and against the cheek. The differences in the perfor-
of 90e100%, where readings were inaccurate but precise for
                                                                                            mance of the pulse oximeter at different attachment sites
‘all-data’ and ‘pass-data’.
                                                                                            highlights the importance of appropriate probe placement
   When compared with cSaO2 (Table 4), the cheek probe gave
                                                                                            when monitoring and assessing oxygen levels in free-ranging
inaccurate but precise readings in the 90e100% range for
                                                                                            wild ungulates during immobilization.
‘pass-data’. Within all the other ranges analyzed, the cheek
                                                                                               The co-oximeter was not available on all study days, which
probe gave inaccurate and imprecise readings.
                                                                                            resulted in a smaller sample size of SaO2 measurements when
Discussion                                                                                  compared with cSaO2 readings. However, the number of
                                                                                            samples was still adequate for the comparisons made, and
In the immobilized impala, SpO2 measurements using a com-                                   similar findings were obtained when SpO2 values were
mercial veterinary pulse oximeter device were accurate and                                  compared with cSaO2. Many studies also compare the pulse
precise indicators of oxygenation. However, it was only                                     rate measured by the pulse oximeter with the heart rate (HR)
acceptable at high SaO2 and was influenced by the probe’s                                    of the animal to determine the quality of the data measured. If
location on the impala. Pulse oximetry was most accurate and                                the two values are similar, this indicates that the pulse oxim-
precise when the probe was placed at the base of the tail and                               eter is measuring a pulse and not movement artefact
the saturation values were greater than 90%. The tail probe                                 (Bohnhorst et al. 2002; Young et al. 2002). However, normal

328        © 2020 Association of Veterinary Anaesthetists and American College of Veterinary Anesthesia and Analgesia. Published by Elsevier Ltd. All rights
                                                                                                                                 reserved., 47, 323e333
Reliability of pulse oximetry in impala TK Mtetwa et al.

Table 1 The number of time-matched paired data obtained from the co-oximeter (SaO2) and the pulse oximeters (SpO2) at the different
attachment sites (ear, cheek, tail and rectum). The number of pairs for all the data (all-data) and their pulse quality are presented. Number of
data pairs excluded, and the criteria of exclusion and the total number of data pairs not excluded (pass-data) are also presented. SD, standard
deviation. sample size (n).

    Attachment site     All-data n     Pulse quality indicator        Pass-data n (%)       Exclusion criteria n (%)
                                       light colour (%)
                                                                                            Triplicate SpO2 with SD >3*          Red light (poor pulse quality)y

    Tail                167            Green: 98                      124 (74)              43 (26)                              0 (0)
                                       Amber: 2
                                       Red: 0
    Ear                 134            Green: 75                      91 (68)               35 (26)                              8 (6)
                                       Amber: 21
                                       Red: 4
    Rectum              153            Green: 83                      115 (75)              37 (24)                              1 (1)
                                       Amber: 16
                                       Red: 1
    Cheek               111            Green: 88                      88 (79)               22 (20)                              1 (1)
                                       Amber: 11
                                       Red: 1

*
 Triplicate SpO2 data excluded because the SD was more than 3% between the SpO2 triplicate data, even though the pulse quality indicated a green or amber light for the
readings.
y
 Triplicate SpO2 data excluded because one or more SpO2 readings had a poor pulse quality indicated by a red light.

variability between HR measured from an electrocardiograph                            data was based solely on the pulse quality determined by the
and the pulse rate measured by pulse oximetry have been                               internal light emitting diode function and the variability of the
observed in cats, dogs and horses (Mathews et al. 2003).                              data. In some instances, more than 30% of good pulse quality
Therefore, using HR alone to assess data quality can have its                         data were excluded because they were highly variable, possibly
limitations. Most studies that validated their readings using                         indicating an error in the measurement of pulse quality, or
pulse rate did not take pulse quality, or its variability, into                       acute changes in SaO2 of the animal at the time of measure-
consideration, and this practice may limit the ability to make                        ment or movement.
valid comparisons. We were unable to record the HR at the                                Previously, the accuracy of pulse oximetry has been assessed
same time as our data collection. Therefore, the quality of our                       by either comparison of SpO2 values with calculated SaO2

Table 2 The number of time-matched paired data obtained from the blood gas analyser (cSaO2) and the pulse oximeters (SpO2) at the different
attachment sites (ear, cheek, tail and rectum). The number of pairs for all the data (all-data) and their pulse quality are presented. Number of
data pairs excluded, and the criteria of exclusion and the total number of data pairs not excluded (pass-data) are also presented. SD, standard
deviation. sample size (n).

    Attachment site     All-data n     Pulse quality indicator       Pass-data n (%)        Exclusion criteria n (%)
                                       light colour (%)
                                                                                            Triplicate SpO2 with SD >3*          Red light (poor pulse quality)y

    Tail                194            Green: 97                     134 (69)               60 (31)                              0 (0)
                                       Amber: 3
                                       Red: 0
    Ear                 139            Green: 72                     80 (58)                50 (36)                              9 (6)
                                       Amber: 24
                                       Red: 4
    Rectum              176            Green: 80                     129 (73)               46 (26)                              1 (1)
                                       Amber: 19
                                       Red: 1
    Cheek               136            Green: 87                     105 (77)               30 (22)                              1 (1)
                                       Amber: 12
                                       Red: 1

*
 Triplicate SpO2 data excluded because the SD was more than 3% between the SpO2 triplicate data, even though the pulse quality indicated a green or amber light for the
readings.
y
 Triplicate SpO2 data excluded because one or more SpO2 readings had a poor pulse quality indicated by a red light.

© 2020 Association of Veterinary Anaesthetists and American College of Veterinary Anesthesia and Analgesia. Published by Elsevier Ltd. All rights                 329
reserved., 47, 323e333
Reliability of pulse oximetry in impala TK Mtetwa et al.

Table 3 The performance of the pulse oximeter readings (SpO2) compared with the gold standard co-oximetry (SaO2) at different attachment
sites in immobilized impala (number of impala ¼ 16). SaO2, oxygen haemoglobin saturation measured by co-oximetry.

  Attachment site         SaO2 Ranges           ‘All-data’                                                      ‘Pass-data’

                                                n         Bias       Precision        Arms       LOA            n        Bias       Precision        Arms       LOA

  Tail                    0e100%                167       6          7                10         21, 8         124      6          7                9          19, 7
                          70e100%               144       4          5                6          14, 5         110      4          5                6          13, 5
Reliability of pulse oximetry in impala TK Mtetwa et al.

Table 4 The performance of the pulse oximeter readings compared with the blood gas analyser (cSaO2) at different attachment sites in
immobilized impala (number of impala ¼ 16). Arms, area root mean squares; LOA, limits of agreement (maximum and minimum); cSaO2,
oxygen haemoglobin saturation was calculated from arterial partial pressure of oxygen as measured by EPOC blood gas analyser.

    Attachment site    cSaO2 Ranges (%)        ‘All-data’                                                 ‘Pass-data’

                                               n       Bias (%)    Precision (%)      Arms    LOA         n      Bias (%)     Precision (%)     Arms     LOA

    Tail               0e100                   194     6           9                  11      24, 12     134    5            8                 9        20, 10
                       70e100                  157     3*          6                  7       16, 9      113    3*           5                 6        13, 7
Reliability of pulse oximetry in impala TK Mtetwa et al.

placement, Barton et al. (1996) used a sensor specifically                      Authors’ contributions
designed for the rectum and secured it using a soft silicone
                                                                                  TM: preparation of equipment, data collection, statistical
plug, factors that may account for their consistent results.
                                                                               analysis and preparation of manuscript. LM, GZ: idea of the
However, a similar study performed by Giguere et al. (2014)
                                                                               study, data collection, data interpretation and preparation of
also indicated larger bias (23%) at saturations below 85%,
                                                                               manuscript. LL: data collection and edited manuscript. SP: data
when the probe was placed inside the rectal mucosa of
                                                                               collection and edited manuscript.
anaesthetized neonatal foals.
   Etorphine is known to cause severe hypoxaemia in immo-
                                                                               Conflict of interest statement
bilized impala (Meyer et al. 2010; Zeiler et al. 2015). During
immobilization, overestimation of pulse oximeter measure-                         SP and LL work for Wildlife Pharmaceuticals, the company
ments is arguably worse than underestimation, since it may                     that sells the drugs that were used to immobilize the animals in
give the impression that an animal has normal oxygen levels                    this study. The aim of this study was not to determine the ef-
when it is hypoxaemic in reality. In this study, probes that                   fects of these drugs and the authors do not advocate their
were placed in the rectum and on the cheek overestimated                       specific use in this manuscript. The other authors declare no
SaO2, therefore these sites should be avoided. This fact is                    conflict of interest.
particularly important especially in clinically ill animals.
                                                                               References
Conclusion
                                                                               Barton LJ, Devey JJ, Gorski S et al. (1996) Evaluation of trans-
We have shown that pulse oximetry, using a commercial                            mittance and reflectance pulse oximetry in a canine model of
veterinary device, is valid in immobilized impala, but only                      hypotension and desaturation. J Vet Emerg Crit Care 6, 21e28.
when oxygen haemoglobin saturations are high (>90%) and                        Bardell D, West W, Senior JM (2017) Evaluation of a new handheld
when a transflectance probe is used at the base of the tail. Since                point-of-care blood gas analyser using 100 equine blood samples.
chemical immobilization often results in severe hypoxaemia in                    Anaesth Analg 44, 1e9.
impala, the interpretation of pulse oximetry readings to mea-                  Batchelder PB, Raley DM (2007) Maximizing the laboratory setting
sure haemoglobin oxygen saturation should be done with                           for testing devices and understanding statistical output in pulse
caution. The authors recommend the use of additional moni-                       oximetry. Anaesth Analg 105, S85eS94.
                                                                               Bland JM, Altman DG (1986) Statistical methods for assessing
toring equipment such as blood gas analysis or co-oximetry.
                                                                                 agreement between two methods of clinical measurements.
However, given that pulse oximetry is cost-effective, nonin-
                                                                                 Lancet 1, 307e310.
vasive and rapidly measures trends in blood oxygenation, the                   Bland JM, Altman DG (2007) Agreement between methods of
development of more reliable pulse oximeters for use in wild                     measurement with multiple observations per individual.
antelope species is warranted. Using the tail base as an optimal                 J Biopharm Stat 17, 571e582.
site for pulse oximetry measurements in impala may help with                   Bohnhorst B, Peter CS, Poets CF (2002) Detection of hyperoxaemia
this development and refinement of future clinical practices.                     in neonates: data from three new pulse oximeters. Arch Dis Child
                                                                                 Fetal Neonatal Ed 87, F217eF219.
Funding                                                                        Burns PM, Driessen B, Boston R, Gunther RA (2006) Accuracy of a
                                                                                 third (Dolphin Voyager) versus first generation pulse oximeter
The study was funded by the University of Pretoria, Faculty of
                                                                                 (Nellcor N-180) in predicting arterial oxygen saturation and pulse
Veterinary Science, Department of Paraclinical Science, South                    rate in the anesthetized dog. Vet Anaesth Analg 33, 281e295.
African Veterinary Association (Wildlife Group) and the Na-                    Chaffin MK, Mathews NS, Cohen ND, Carter GK (1996) Evaluation
tional Research Foundation.                                                      of pulse oximetry in anaesthetised foals using multiple combi-
                                                                                 nations of transducer type and transducer attachment site.
Acknowledgements                                                                 Equine Vet J 28, 437e445.
                                                                               Clerbaux T, Gustin P, Detry B et al. (1993) Comparative study
  The authors would like to thank Wildlifevets.com and                           of the oxyhaemoglobin dissociation curve of four mammals:
Wildlife Pharmaceuticals for their generous support and help                     man, dog, horse and cattle. Comp Biochem Physiol 106A,
with data collection. Additionally, we thank Dr Etienne Bason                    687e694.
and Penelope Miya for their help. We also thank Kyron                          Coghe J, Uystepruyst C, Bureau F, Lekeux P (1999) Non-invasive
Laboratories and Ascendis Medical for loaning the Nonin                          assessment of arterial haemoglobin oxygen saturation in cattle
pulse oximeters and Co-oximeter devices, and the University                      by pulse oximetry. Vet Rec 145, 666e669.
of Pretoria, Faculty of Veterinary Science, Department of                      Dawson J, Bastrenta P, Cavigioli F et al. (2014) The precision and
Paraclinical Science and National Research Foundation for                        accuracy of Nellcor and Masimo oximeters at low oxygen satu-
                                                                                 rations (70%) in newborn lambs. Arch Dis Child Fetal Neonatal
funding.
                                                                                 Ed 99, F278eF281.

332       © 2020 Association of Veterinary Anaesthetists and American College of Veterinary Anesthesia and Analgesia. Published by Elsevier Ltd. All rights
                                                                                                                                reserved., 47, 323e333
Reliability of pulse oximetry in impala TK Mtetwa et al.

DeMeulenaere S (2007) Pulse oximetry: uses and limitations.                      Reiners JK, Rossdeutsher W, Hopster K et al. (2018) Development
  J Nurs Pract 3, 312e317.                                                         and clinical evaluation of a new sensor design for buccal pulse
Fahlman A (2014) Oxygen therapy. In: Zoo Animal and Wildlife                       oximetry in horses. Equine Vet J 50, 228e234.
  Immobilization and Anesthesia (2nd edn). West G, Heard D,                      Serfontein GL, Jaroszewicz AM (1978) Estimation of gestational age
  Caulkett N (eds). John Wiley & Sons, Inc., NJ, USA. pp. 69e81.                   at birth: comparison of two methods. Arch Dis Child 53, 59e511.
Grap MJ (2002) Pulse oximetry. Critical Care Nurse 22, 69e74.                    Uystepruyst CH, Coghe J, Bureau F, Lekeux P (2000) Evaluation of
Giguere S, Sanchez LC, Shih A (2014) Accuracy of calculated arterial               accuracy of pulse oximetry in newborn calves. Vet J 159, 71e76.
  saturation in oxygen in neonatal foals and effect of monitor, sensor,          Quinn CT, Raisis AL, Musk GC (2013) Evaluation of Masimo signal
  site of sensor placement, and degree of hypoxemia on the accuracy                extraction technology pulse oximetry in anaesthetized pregnant
  of pulse oximetry. J Vet Emerg Crit Care 25, 539.                                sheep. Vet Anaesth Analg 40, 149e156.
Grubb TL, Anderson DE (2017) Assessment of clinical application of               van de Louw A, Cracco C, Cerf C et al. (2001) Accuracy of pulse
  pulse oximetry probes in llamas and alpacas. Vet Med Sci 3,                      oximetry in the intensive care unit. Intens Care Med 27,
  169e175.                                                                         1606e1613.
Haymerle A, Knauer F, Walzer C (2016) Two methods to adapt the                   Wong DM, Alcott CJ, Wang C et al. (2011) Agreement between
  human haemoglobin-oxygen dissociation algorithm to the blood                     arterial partial pressure of carbon dioxide and saturation of he-
  of the white rhinoceros (Ceratotherium simum) and to determine                   moglobin with oxygen values obtained by direct arterial blood
  the accuracy of pulse oximetry. Vet Anaesth Analg 43, 566e570.                   measurements versus noninvasive methods in conscious healthy
ISO 80601-2-61:2011 DIN EN (2011) Medical electrical equip-                        and ill foals. J Am Vet Med Assoc 239, 1341e1347.
  ment- Part 2-61: particular requirements for basic safety and                  Whitehair KJ, Watney GCG, Leith DE, Debowes RM (1990) Pulse
  essential performance of pulse oximeter equipment (ISO 80601-                    oximetry in horses. Vet Surg 19, 243e248.
  2-61:2011). German version EN ISO 80601-2-61:2011.                             Young SS, Skeans SS, Lamca JE, Chapman RW (2002) Agreement
Lee WW, Mayberry K, Crapo R, Jensen RL (2000) The accuracy of                      of SpO2, SaO2 and ScO2 in anesthetized cynomolgus monkeys
  pulse oximetry in the emergency care unit. Am J Emerg Med 18,                    (Macaca fascicularis). Vet Anaesth Analg 29, 150e155.
  427e431.                                                                       Zeiler GE, Meyer LCR (2017a) Comparison of thiafentanil-
Martin-Jurado O, Bektas R, Fahrton A et al. (2011) Comparison of                   medetomidine to etorphine-medetomidine immobilization of im-
  the effects of racemic ketamine and s-ketamine for anesthesia in                 palas (Aepyceros melampus). J S Afr Vet Assoc 88, 1e8.
  rheem gazelles (Gazella subgutturosa marica) and subgutturosa                  Zeiler GE, Meyer LCR (2017b) Captive management of wild impala
  gazelles (Gazella subgutturosa subgutturosa). Am J Vet Res 72,                   (Aepyceros melampus) during intensive immobilization and
  1164e1170.                                                                       anaesthesia study trials. J Zoo Wildl Med 48, 1058e1071.
Mathews NS, Hartke S, Allen JC (2003) An evaluation of pulse                     Zeiler GE, Stegmann GF, Fosgate G et al. (2015) Etorphi-
  oximeters in dogs, cats and horses. Vet Anaesth Analg 30, 3e14.                  neeketamineemedetomidine total intravenous anaesthesia in
Meyer LCR, Hetem RS, Fick LG et al. (2008) Thermal, cardiore-                      wild impala (Aepyceros Melampus) of 120-minute duration. Vet
  spiratory and cortisol responses of impala (Aepyceros melampus) to               Anaesth Analg 6, 755e766.
  chemical immobilization with 4 different drug combinations. J S
  Afr Vet Assoc 79, 121e129.                                                     Received 18 February 2019; accepted 6 August 2019.
Meyer LCR, Hetem RS, Fick LG et al. (2010) Effects of serotonin
                                                                                 Available online 4 February 2020
  agonists and doxapram on respiratory depression and hypoxemia
  in etorphine-immobilized impala (Aepyceros Melampus). J Wildl
  Dis 46, 514e524.
Oldham HG, Bevan MM, McDermott M (1979) Comparison of new
  miniature wright peak flow meter with the standard wright peak
                                                                                 Appendix A
  flow meter. Thorax 34, 807e808.
Pedersen T, Hovhannisyan K, Moller AN (2009) Pulse oximetry for
  perioperative monitoring. Cochrane Database Syst Rev 4,                                    qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
                                                                                              X                                                ffi
  CD002013.                                                                      Arms ¼               ni¼n ðSpO2  SaO2 Þ2 n                                            (1)
Pfitzer S, Laubscher L, Meyer L et al. (2019) Dose-effect study of the
  serotonin agonist R-8-OH-DPAT on opioid-induced respiratory                                qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
                                                                                              Xn                                           
  depression in blesbok (Damaliscus pygargus philipsi) and impala                Arms ¼              i¼n
                                                                                                           ðSpO2  SaO2 Þ2 n
  (Aepyceros melampus). Vet Anaesth Analg 46, 796e806.
                                                                                        rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
                                                                                         X                                                            . 
Proulx J (1999) Respiratory monitoring: Arterial blood gas analysis,
  pulse oximetry, and end-tidal carbon dioxide analysis. Clin Tech               Arms ¼               ði ¼ nÞ^nðSpO2  SaO2 Þ^2 n
  Small Anim Prac 14, 227e230.
                                                                                        sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
                                                                                         Pn                                        2
                                                                                              i¼n ðSpO2  SaO2Þ
                                                                                 Arms ¼
                                                                                                              n

© 2020 Association of Veterinary Anaesthetists and American College of Veterinary Anesthesia and Analgesia. Published by Elsevier Ltd. All rights                       333
reserved., 47, 323e333
You can also read